| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the School of Pharmacy, University of Queensland, Queensland, Australia.
Address correspondence and reprint requests to Peter J. Cabot, PhD, The School of Pharmacy, The University of Queensland, 4072, Queensland, Australia. Address e-mail to pcabot{at}pharmacy.uq.edu.au.
Abstract
BACKGROUND: Opioid-containing immune cells migrate in a site-directed manner into inflamed tissue and adhere to sensory nerve fibers. These cells release opioid peptides in close proximity to these fibers, thereby avoiding localized degradation by peptidases, and delivering opioid peptides proximal to opioid receptors to provide antinociception.
METHODS: The effects of the anti–neural-cell-adhesion molecule (anti-NCAM) were assessed on cold water swim stress-induced antinociception in Wistar rats with Freund’s adjuvant-induced inflammation of one hindpaw. Algesiometry was assessed for both thermal and mechanical stimuli. Cell adhesion experiments examining the effects of ß-endorphin and antibodies to NCAM and intercellular cell adhesion molecule-1 and were performed on cultured dorsal root ganglion neurons and isolated lymphocytes. Lymphocyte binding was determined by fluorescence using calcein AM loaded into freshly isolated lymphocytes.
RESULTS: The direct adhesion between lymphocytes and cultured sensory neurons was inhibited by anti-NCAM. This adhesion was also demonstrated to be opioid dependent, with lymphocyte adhesion to cultured sensory neurons reduced in the presence of 1 µM ß-endorphin, which was reversed by 100 µM naloxone. Moreover, anti-NCAM blocked cold-water-swim-induced analgesia in inflamed paws both to thermal and mechanical stimuli. However, anti-NCAM did not affect fentanyl-induced antinociception.
CONCLUSIONS: This study provides insight into the role of cell adhesion molecules in lymphocyte adhesion to sensory neurons and a link to immune-derived antinociception.
Inflammatory nociception can be controlled by numerous endogenous mechanisms. Immune cells infiltrate into inflamed tissue and release endogenous opioid peptides that interact with opioid receptors on local sensory nerve fibers (1). This interaction leads to potent antinociception that is consistent with the local application of exogenous analgesics (2). These cells migrate into inflamed tissue using adhesion molecules expressed on the surface of the inflamed endothelium and the immune cells (3). Selective blockade of this process has been shown to inhibit the antinociceptive action of immune cell-derived opioids by preventing immune cell access to sites of inflammation (3–5).
Inflammation is characterized by immune cell recruitment and infiltration into the site of injury. These cells have a comprehensive role in both inflammation and control of nociception. Immune cells, particularly lymphocytes, express and store opioid peptides (6). They also release endogenous opioid peptides, such as ß-endorphin, met-enkephalin, and dynorphin-A, on stimulation with corticotropin-releasing factor or the cytokine, interleukin-1ß, in inflamed tissue (1,7) Opioid peptide release brings about an antinociceptive response in the inflamed tissue. However, inflamed tissue is a concentrated milieu of peptidases (8), increased acidity (9), and increased temperatures (10). These factors account for an increased metabolic environment and have been shown to expedite the degradation of opioid peptides in inflamed tissue (11). It is likely that, given this highly metabolic environment, effective antinociception can only be achieved by the release of endogenous opioid peptides proximal to sensory nerve endings in the inflamed tissue. Increased antinociception has been demonstrated in the presence of peptidase inhibitors (12), and peptidases have been shown to inhibit µ opioid receptors (MOR) receptor internalization by endogenous opioids in the dorsal horn (13). Direct adhesion between sensory neurons and lymphocytes within inflamed tissue may be the only way to minimize increased degradation of endogenous opioids and, therefore, ensure adequate quantities at local opioid receptors to elicit antinociception. Indeed, immune cells have been shown to interact directly with adhesion molecules on neurons (14). Ligands for cell adhesion molecules, such as neural cell adhesion molecule (NCAM), are expressed on both immune cells and neurons and have been shown to be involved in the interaction between the immune system and the central nervous system (15). More recently, comparisons have been made between neuronal and immunological synapses (16). Therefore, in this study, we evaluated direct cell adhesion between immune cells and peripheral sensory neurons and investigated the link to peripheral antinociception.
METHODS
Subjects
Experiments were conducted in male Wistar rats (Central Animal Breeding House, University of Queensland), 200–350 g in body weight, housed individually, with standard rodent chow and water available ad libitum. Room temperature was maintained at 22°C, and a 12/12 h light–dark cycle used. All animal experimentation in this study was performed with permission from the Animal Experimentation and Ethics Committee, University of Queensland.
Reagents
Freund’s complete adjuvant (FCA), anti-ß-endorphin polyclonal IgG, ß-endorphin peptide, Hank’s buffered salt solution (HBSS), porcine trypsin, collagenase, DNAse, trypsin-chymotrypsin inhibitor, horse serum, fetal calf serum, nerve growth factor (NGF), poly-d-lysine, PAP pen, bovine serum albumin (BSA), trypan blue, gelatin, and penicillin/streptomycin were obtained from Sigma–Aldrich (Sydney, Australia). Goat anti-rat NCAM-L1 was from Santa Cruz Biotechnology (Santa Cruz, CA), and mouse anti-rat intercellular cell adhesion molecule-1 (ICAM-1) monoclonal antibody was from Chemicon (Chemicon International, Temecula, CA). Rabbit anti-goat IgG conjugated with FITC was obtained from Bioscientific (Gymea, Australia). CD4 anti-rat monoclonal IgG conjugated with PE and Lympholyte®-Mammal were obtained from Cedarlane Laboratories (Hornby, Canada). Fentanyl (30 µg/mL) was purchased from the Royal Brisbane Hospital (Brisbane, Australia). Dulbecco’s modified eagle medium (DMEM), neurobasal- A medium, B27, and heparinized saline were from Invitrogen Life Technologies (Mount Waverly, Australia). Isoflurane was from Abbot Australia (Cronulla, Australia), and the SlowFade® Antifade kit was purchased from Molecular Probes (Eugene, OR).
Induction and Evaluation of Inflammation
Rats sedated by brief isoflurane exposure received an intraplantar injection (i.pl.) of 150 µL FCA into the right hindpaw. Concurrent with FCA injection, a 1:100 dose of anti-NCAM within 100 µL HBSS or HBSS was administered via i.pl injection to rats; the dose was determined based on preliminary dosing studies. Inflammation was measured with a paw volume plethysmometer (Ugo Basile, Comerio, Italy). Nociceptive thresholds were assessed using both the paw pressure analgesiometer (Ugo Basile, Comerio, Italy) and the paw thermal analgesiometer (Ugo Basile, Comerio, Italy).
Algesiometry
Two days before the studies, rats were handled twice by an experimenter. Baseline measurements of paw pressure threshold, paw thermal threshold, and paw volume were obtained. Six hours after concurrent anti-NCAM (40 µg/kg) and FCA administration, paw volume, baseline paw pressure, and thermal thresholds were assessed. Anti-NCAM effects were examined on antinociception produced from cold-water-swim stress (CWS), elicited from 1 min immersion in 4°C water and assessed 5 min after the CWS. During experiments, the sequence of inflamed and noninflamed paw testing was alternated to preclude order effects for both paw pressure threshold and paw thermal threshold with approximately 5 s between measurements. Cut offs were 250 g for pressure threshold and 20 s for thermal threshold.
Anti-NCAM Effects on Fentanyl-Induced Antinociception
To assess the effects of anti-NCAM on opioid-induced antinociception rats inoculated with FCA and 40 µg/kg of anti-NCAM where injected with i.pl. 100 µL fentanyl (30 µg/mL) at 6 h after FCA. Antinociceptive effects were determined using paw pressure threshold at 5 min after i.pl. fentanyl.
Dorsal Root Ganglion Cell Culture
Dorsal root ganglion (DRG) neurons were collected and cultured from adult male Wistar rats (200–350 g). In brief, DRG neurons were collected from the rat spinal cord under sterile conditions from level T12 to S2. The harvested DRGs were placed immediately in cold DMEM (supplemented with 4 mM glutamine). After transfer to DMEM containing trypsin (0.5 mg/mL), collagenase (1 mg/mL) and DNAse (1 µg/mL), the DRGs were minced with surgical scissors, agitated 15–25 times with a fire-polished Pasteur pipette, and incubated at 37°C and 5% CO2 for 1 h. Trypsin-chymotrypsin inhibitor was then added, and the cells were again agitated 15–25 times with a fire-polished Pasteur pipette. The cell suspension was centrifuged at 200 g for 10 min, the supernatant was removed, and the cell pellet resuspended in 2 mL of warm medium consisting of DMEM, 10% horse serum, 10% fetal bovine serum, penicillin (2 mM), streptomycin (2 mM), and NGF (25 ng/mL). The cell count was obtained using a hemacytometer, and the cells were plated on 96-well glass bottom plates coated with poly-d-lysine (0.05 mg/mL) at a density of 4 x 105 cells per well. The cells were kept undisturbed at 37°C in a 5% CO2 incubator for 1 h. The media was then changed to neurobasal-A medium supplemented with glutamine (4 mM), NGF (25 ng/mL), and B27, to reduce glial cell growth. The cells were again fed with this neurobasal-A medium after 24 h and studied after 3 days in culture.
Isolation of Lymphocytes from Rat Blood
Lymphocytes were harvested from nontreated adult male Wistar rats. The procedure involves lightly anesthetizing rats with 50% CO2/O2. The rats were then decapitated, and trunk blood was collected (7–15 mL) into tubes containing 3 mL heparinized saline. The blood (excluding the 3 mL heparin) was diluted with equal volume of HBSS (1:1), and 4 mL of this suspension was layered over 3 mL of Lympholyte®-Mammal. After centrifugation (800 g, 20 min, 24°C) the lymphocyte layer was carefully removed and resuspended in HBSS (1:1) to reduce the density of the solution. Samples were centrifuged (800 g; 10 min), the supernatant was decanted, and the pellet reconstituted with 2–4 mL of HBSS. An aliquot (20 µl) of the suspension was removed for lymphocyte counts and determination of cell viability (>95%) using the Trypan blue exclusion method.
Cell Adhesion Between Lymphocytes and Cultured DRG Neurons
Pretreatment of lymphocytes and DRG with either ß-endorphin (1 µM), anti-ICAM-1 (1:100), and anti-NCAM (1:100) or control (physiological buffer solution [PBS]) was performed by incubating DRG neurons for 2 h before the loading of cells and was included in the respective loading buffers to ensure incubation was constant for both lymphocytes and DRG neurons before and during adhesion experiments. Lymphocytes collected were centrifuged for 5 min at 5000 rpm. The supernatant was then removed, and the lymphocytes were loaded with calcein AM (5 µM) for 30 min at pH 7.2 at room temperature. The lymphocytes were then washed with PBS and centrifuged for 5 min at 5000 rpm to remove excess calcein AM. After centrifugation, the supernatant was removed and the lymphocytes were resuspended in PBS, before being incubated with the cultured DRG neurons at a density of 150,000 cells per well for 30 min at room temperature. The wells were washed two to three times with PBS to remove nonadherent lymphocytes; each well was made up to exactly 100 µL of PBS, and the fluorescence was read on the NovostarTM fluorescent microplate reader. Adherent cells were spun off at 10,000 rpm, and plates were read to determine nonspecific binding. The data were presented as percentage cell binding relative to control binding experiments on each plate and adjusted for nonspecific binding. Fluorescent readings were compared to digital images taken using fluorescence microscopy (Meta-FluorTM).
Immunolabeling for Cell Counting in Inflamed Paw Tissue
Frozen stored hindpaw tissue was embedded into OCT compound 4583 (Tissue Tek®), snap-frozen, and 10-µm coronal sections were cut using a Leica cryostat®. Sections were lifted onto gelatin-coated slides, and immunolabeling was performed as described previously (17). Hydrophobic wells were made around cryostat tissue sections with a PAP pen, tissue fixed with acetone (10 s, 4°C), and then washed with PBS (pH 7.4) (10 min). To block nonspecific reactive groups, sections were incubated in 0.5% BSA-PBS for 30 min. Between each incubation step, the sections were washed three times with PBS for 15 min. Antibodies were applied to tissue sections and incubated in a humidity chamber for the following durations; rabbit anti-ß-endorphin at 1:800 PBS-BSA (16 h, 4°C); anti-rabbit IgG conjugated with FITC at 1:200 PBS-BSA (2 h, 25°C); and preconjugated monoclonal anti-CD4 with PE at 1:250 PBS-BSA (2 h, 25°C). Sections were covered with cover slips with the aid of a SlowFade® Antifade kit and viewed using a fluorescence microscope with Meta-morph® and Metafluor® imaging software. Specificity of polyclonal antibody staining was verified by preabsorption of anti-ß-endorphin with ß-endorphin (10 µm, 24 h, 4°C), followed by the same immunolabeling procedure as outlined earlier. Cell counting was performed on at least three sections from six individual rats for each group over a section area of 110 mm2.
Statistical Analysis
All data are expressed as means ± sem. Comparisons were made using the Wilcoxon’s matched pairs test for dependent data, and the Mann–Whitney U-test for independent data. Dose-relationship curves were examined using analysis of variance followed by a linear regression analysis. Differences were considered significant if P < 0.05.
RESULTS
Cell Adhesion Between Lymphocytes and Cultured DRG
The role of cell adhesion molecules on the direct adhesion between lymphocytes and cultured DRG neurons was assessed after pretreatment with anti-NCAM (1:100), anti-ICAM-1 (1:100), anti-NCAM (1:100) + anti-ICAM-1 (1:100), and control. Cell adhesion molecule blockade with ICAM-1 (P < 0.05), NCAM (P < 0.01), and both NCAM + ICAM-1 (P < 0.01) significantly reduced the degree of cell adhesion between lymphocytes and DRG neurons compared with control (Fig. 1). Control binding represented binding of 5.6% of total cells added to each well.
|
Preincubation with 1 µM ß-endorphin (n = 8) significantly decreased cell binding between lymphocytes and cultured DRG neurons compared to control (Fig. 2, P < 0.05). This effect was completely reversed after treatment with 100 µM naloxone (n = 8, P < 0.05). The use of naloxone demonstrates that the effect of ß-endorphin on attenuating cell binding between lymphocytes and cultured DRG neurons is opioid receptor-dependent. Treatment with 100 µM naloxone alone (n = 7) did not significantly alter the percentage of cell binding compared with the control group (Fig. 2, P > 0.05). Fluorescent images of the specific binding were verified by fluorescent microscopy, which showed binding occurred predominantly to neuronal cell bodies and processes (Fig. 3).
|
|
|
Antinociception from CWS blocked by anti-NCAM
CWS produced antinociception assessed as paw thermal thresholds in the inflamed paws of control rats (P < 0.05, Wilcoxon test), but not in rats treated with anti-NCAM (40 µg/kg, 100 µL) directly to the inflamed paw (Fig. 4A, P > 0.05, Wilcoxon test). Paw thermal thresholds were not significantly altered in the contralateral paws of untreated and anti-NCAM-treated rats (9.99 ± 0.71 and 10.4 ± 0.54 for pre- and post-CWS untreated, 9.76 ± 0.39 and 10.58 ± 0.53 for pre- and post-CWS anti-NCAM-treated, respectively, P < 0.05). Mechanical nociceptive thresholds were similarly reduced in anti-NCAM-treated rats. CWS produced antinociception in the inflamed paws of untreated rats (P < 0.05 Wilcoxon test), but not in rats treated with anti-NCAM directly to the inflamed paw (P > 0.05, Wilcoxon test) (Fig. 4B). Mechanical nociceptive thresholds were not altered in the contralateral paws of untreated and anti-NCAM-treated rats (160 ± 10.5 and 162.71 ± 9.40 for untreated pre- and post-CWS, and 142 ± 9.2 and 152 ± 11.9 for pre- and post-CWS anti-NCAM-treated rats, respectively). Paw volume did not differ between anti-NCAM and sham-injected groups (3.54 ± 0.17 mL and 3.47 ± 0.33 mL, respectively).
Effect of Anti-NCAM on Fentanyl-Induced Antinociception
The decrease in antinociception with anti-NCAM treatment could have been explained by a nonspecific functional effect on opioid receptor function or antinociceptive pathways. To account for this possibility, we delivered 100 µl i.pl. of the opioid agonist fentanyl (30 µg/mL) to induce an antinociceptive response in peripheral tissue. Fentanyl produced similar levels of antinociception in both anti-NCAM and untreated rats for both thermal (Fig. 5A) and mechanical nociceptive assessment (Fig. 5B).
|
Effect of Anti-NCAM on ß-Endorphin-Containing Cell Numbers in Inflamed Tissue and Inflammation
The effect of anti-NCAM on the numbers of CD4+/ß-endorphin cells in inflamed paw tissue was examined. Control paw tissue showed 32 ± 11 and compared to 27 ± 8 cell counts in anti-NCAM inflamed rat paw sections. These data are expressed as mean ± sem (n = 6 rats) (P > 0.05, Mann–Whitney U-test) and the field for counting was represented by 110 mm2 (Fig. 6.). Preabsorption with 10 µM ß-endorphin abolished specific immunolabeling in control sections.
|
DISCUSSION
Immune cells express, contain, and release opioid peptides within inflamed tissue. Local proinflammatory mediators, such as interleukin-1ß and corticotropin-releasing factor, are potent and receptor-specific activators of opioid peptide release (1). The release of these immune-derived opioid peptides produces antinociceptive responses in inflamed tissue that are opioid receptor-specific and can be blocked by antibodies against ß-endorphin, met-enkephalin, and dynorphin (7). Opioid-containing immune cells migrate specifically to inflamed tissue, and inhibition of migration through selectin blockers has been shown to reversibly reduce immune cell infiltration as well as immune-derived antinociception (5,3). Furthermore, immunocytes were shown to co-express ß-endorphin and adhesion molecules within inflamed tissue (18). Inflammation is characterized by a complex milieu of proinflammatory mediators as well as enzymatic peptidases, which when inhibited enhance opioid peptide antinociception (19). It is evident that site-directed infiltration enables passage of immune cells that specifically target inflamed tissue. Consistent with this is the notion that there may be a mechanism by which immune cells migrate specifically to peripheral sensory nerve fibers and bind specifically to these endings. Upon interaction or adhesion with the nerve fibers, opioids peptides may then be released within the effective range of opioid receptors to produce antinociception. This study demonstrates that antibody treatment of adhesion molecules, evident on both immune cells and peripheral sensory nerve fibers, blocks both immune-derived analgesia and the direct binding interaction between cultured DRG and isolated lymphocyte. These findings provide further pharmacological evidence of a role for a neuronal-immune dynamic synapse (16). There is evidence demonstrating an interaction between neuronal and immune tissues in a number of organs, e.g., spleen (20) and airways (21). However, this is the first study to propose a link between immune cells and DRGs and a link to antinociceptive mechanisms.
NCAMs are expressed on the surface of immune cells and neurons (22). These molecules are a member of the immunoglobulin superfamily and are involved in neuronal development, as well as nerve sprouting and regrowth, in response to nerve ligation or injury (23). NCAM has been shown on DRGs (24,25) and is thought to bind by both homophilic and heterophilic interaction (23). Moreover, NCAM has been shown to mediate adhesion between immune cells (26). ICAMs are expressed on the surface of immune cells, endothelium and on neurons (27). ICAM-1 is particularly important in inflammation, immunocyte, and endothelium.
Evidence for a direct interaction between DRG neurons and lymphocytes through NCAM was demonstrated in direct adhesion experiments in our studies. This adhesion was, in part, opioid-mediated, as shown by the effects of ß-endorphin on adhesion between cultured DRG and lymphocytes, and indicating the existence of adhesion mechanisms which may involve opioid control. This may highlight an additional role for opioids in peripheral analgesia, and will be investigated in future studies.
In previous studies, immune cells have been shown to bind directly to neurons in culture (14). Direct adhesion between hippocampal neurons and T lymphocytes has been blocked by antibodies for cell adhesion molecules such as ICAM-5, as well as peptide fragments for this adhesion molecule (14). Consistent with this notion, we examined the effect of anti-NCAM pretreatment on cultured DRG neurons incubated with rat lymphocytes. Our studies demonstrated that anti-NCAM and anti-ICAM-1 blocked the direct cell adhesion between lymphocytes and cultured DRGs.
Immune-derived antinociception was produced by exposing rats to CWS, which releases opioid peptides from immune cells within inflamed tissue (1) and consequentially produces potent opioid receptor-mediated antinociception (4,2). It is evident that cell adhesion molecules play an integral role in immune cell interaction with peripheral neurons, and appear to have a determining influence on the overall antinociceptive outcome of immune cell-derived peptides in inflamed tissue. Anti-NCAM appears to block immune-derived antinociception without altering immune cell infiltration, which suggests that NCAM involvement in immune-derived antinociception is independent of infiltration of immune cells to inflamed tissue seen in previous studies for selectin blockade (3) and ICAM-1 (28). However, electron microscopy studies will be required to elucidate the localization of NCAM-binding sites in neuroimmune binding. Importantly, anti-NCAM pretreatment was shown not to alter opioid antinociception function in vivo by interfering with neuronal function. This was examined by using i.pl. fentanyl to induce antinociception in both anti-NCAM-treated and untreated rats, whereby the difference would only be associated with a nonspecific or selective effect of the antibody on opioid receptor or neuronal function. In addition, based on the opioid-inhibitory effects on lymphocyte adhesion to DRG in culture, fentanyl will likely also interrupt direct adhesion in vivo, altering endogenous opioid-mediated release and, presumably, antinociception. This effect cannot be studied because of the opioid receptor effects of fentanyl on the inflammatory nociception.
This study provides new insights into the interrelationship between the immune system and the peripheral nervous system. Indeed, a range of new targeted therapeutics could use the binding properties of the ICAM-1 or NCAM family to enhance peripheral antinociception, thereby reducing unwanted side effects commonly associated with opioids. Thus, it appears that the site-directed delivery of therapeutic agents is a commonality in endogenous pain mechanisms through a site-directed delivery of antinociception to peripheral sensory nerve targets.
Footnotes
Accepted for publication August 15, 2006.
Supported by the National Health and Medical Research Council.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
